Adaptive Distributed Resampling Algorithm with Non-Proportional Allocation

TL;DR

This paper introduces an adaptive distributed resampling algorithm that dynamically adjusts particle exchange and topology, significantly enhancing performance and convergence speed in parallel particle filtering applications.

Contribution

The paper proposes an adaptive RNA algorithm that improves upon the original by dynamically tuning exchange ratios and topology, leading to faster convergence and better performance.

Findings

ARNA improves runtime performance by about 9% over RNA.

ARNA achieves approximately 20-fold faster convergence.

ARNA requires minimal modifications to existing RNA implementations.

Abstract

The distributed resampling algorithm with proportional allocation (RNA) is key to implementing particle filtering applications on parallel computer systems. We extend the original work by Bolic et al. by introducing an adaptive RNA (ARNA) algorithm, improving RNA by dynamically adjusting the particle-exchange ratio and randomizing the process ring topology. This improves the runtime performance of ARNA by about 9% over RNA with 10% particle exchange. ARNA also significantly improves the speed at which information is shared between processing elements, leading to about 20-fold faster convergence. The ARNA algorithm requires only a few modifications to the original RNA, and is hence easy to implement.